The proposed research involves six specific subject areas: (a) survival analysis aims at continuing the development of statistical procedures appropriate for the analysis of censored survival data; (b) growth curves analyses are appropriate for laboratory or clinical longitudinal studies when response data may be missing or heterogeneous; (c) tests for multivariate normality will be valuable for characterizing multivariate data sets and assessing whether common parametric multivariate data-analytic procedures may safely be applied to them; (d) the development of mathematical models for monoclonal antibody-tumor systems is of importance to immunologists and oncologists, and will be applicable to the staging of therapeutic monoclonal antibody preparations in various malignant settings; (e) statistical considerations in the design and analysis of in vitro drug testing techniques based on the human tumor colony forming assay will be addressed, in order to enhance the applicability of such techniques in clinical oncology; (f) statistical issues relevant to the design of Phase I, Phase II, and Phase III trials in clinical oncology will be studied.